Skip to content

Module for CandyBot to help recognise people and objects via web-cam

Notifications You must be signed in to change notification settings

FunRobots/candybot_v2

 
 

Repository files navigation

  1. Setup

    1.1. install docker image candybot_vr

     sudo docker pull funrobots/candybot_vr
    

    1.2. install python3 libraries in docker:

     pip3 install luma.core git+https://github.com/adafruit/Adafruit_Python_PCA9685.git
    

    1.3. Run candybot_vr image

    sudo docker run -ti --privileged --device /dev:/dev candybot_vr

    1.4. Run candybot_v2 ROS package (in docker container):

     1.4.1. create catkin workspace:
     
     		mkdir -p ~/catkin_ws/src
     	
     		cd ~/catkin_ws
     	
     		catkin_make
    
     1.4.2. clone candybot_v2 package:
     		
     		cd ~/catkin_ws/src
     
     		git clone https://github.com/FunRobots/candybot_v2.git
     
     1.4.3. run candybot_v2 package:
    
     		cd ~/catkin_ws
     
     		rosrun candybot_v2 run.py
    
     1.4.4. run just one candybot_v2 package node:
     
     		roscore
     
     		in another terminal (see point 3):
     	
     			cd ~/catkin_ws
     	
     			source devel/setup.bash
     	
     			rosrun candybot_v2 <node_name>
    

    1.5. save changes in docker image (commit):

     in Raspbian terminal:
     	
     	- list containers and copy candybot_vr id:
     		
     		sudo docker container ls
     		
     	- commit changes:
     		
     		sudo docker commit <candybot_id> <your_image_name>
     		
     			(you can save source container locally with any name)
    
  2. Open new terminal in running candybot_vr docker container:

    in new window (for example, in new browser tab with dataplicity.com):

     sudo docker exec -ti candybot_vr /bin/bash
    
  3. Run tests (in running docker container candybot_vr):

    1. in the first terminal run roscore

    in other terminals:

    1. go into folder catkin_ws:

      cd ~/catkin_ws

    2. export packages setups:

      source devel/setup.bash

    3. to run tests:

      3.1.сommon way to run test:

       	rosrun candybot_v2 <test_file_name>
      

      3.2. some tests require the tested module to be started

       in this case:
      
       1. run the test module (for example, motion_<part>_controller.py)
      
       2. in another terminal (see 3.) do 4.1 - 4.3.1
      

About

Module for CandyBot to help recognise people and objects via web-cam

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published